The role of fruit bats in the transmission of pathogenic leptospires in Australia.

Although antileptospiral antibodies and leptospiral DNA have been detected in Australian fruit bats, the role of such bats as infectious hosts for the leptospires found in rodents and humans remains unconfirmed. A cohort-design, replicated survey was recently conducted in Far North Queensland, Australia, to determine if the abundance and leptospiral status of rodents were affected by association with colonies of fruit bats (Pteropus conspicillatus spp.) via rodent contact with potentially infectious fruit-bat urine. In each of four study areas, a 'colony site' that included a fruit-bat colony and the land within 1500 m of the colony was compared with a 'control site' that held no fruit-bat colonies and was >2000 m from the nearest edge of the colony site. Rodents were surveyed, for a total of 2400 trap-nights, over six sampling sessions between September 2007 and September 2008. A low abundance of rodents but a high carriage of leptospires in the rodents present were found to be associated with proximity to a fruit-bat colony. For example, means of 0·4 and 2·3 fawn-footed melomys (Melomys cervinipes) were collected/100 trap-nights at sites with and without fruit-bat colonies, respectively (P<0·001), but the corresponding prevalences of leptospiral carriage were 100% and 3·6% (P<0·001). Such trends were consistent across all of the sampling sessions but not across all of the sampling sites. Leptospires were not isolated from fruit bats by culture, and the role of such bats in the transmission of leptospires to rodents cannot be confirmed. The data collected do, however, indicate the existence of a potential pathway for transmission of leptospires from fruit bats to rodents, via rodent contact with infectious fruit-bat urine. Fruit bats may possibly be involved in the ecology of leptospires (including emergent serovars), as disseminators of pathogens to rodent populations. Stringent quantitative risk analysis of the present and similar data, to explore their implications in terms of disease prevalence and wildlife population dynamics, is recommended.